Data Science Projects for Freelancers

I’m a computer engineering student pulling together a project that hinges on clear, well-presented descriptive analytics. I already have the raw dataset ready to share; what I’m missing is an accurate, insight-rich summary that I can confidently submit ahead of my deadline.

Here’s what I need from you:
• Clean the data if required and document any assumptions.
• Produce core descriptive statistics (mean, median, mode, range, standard deviation, etc.).
• Create straightforward visualisations—think histograms, box plots, bar charts—that instantly communicate the story in the numbers.
• Compile a concise report (PDF or DOCX) explaining the methods you used, the insights uncovered, and any noteworthy anomalies.
• Provide the reproducible code or spreadsheet formulas (Python with pandas/matplotlib, R with tidyverse, or Excel—use whichever you’re strongest with).

Accuracy and clarity are paramount; I would like the full package delivered ahead of the agreed hand-in so I have time for review. If this sounds like a straightforward spin-up of your data analysis skills, I’m ready to get the dataset over to you and kick things off right away.

Skills: Python, Statistics, Statistical Analysis, SPSS Statistics, Data Science, Data Visualization, Data Analysis, Pandas

We are looking for a researcher to produce an original, TRIPOD+AI-compliant prediction finding using NHANES, a large, nationally representative US health survey that links questionnaire, examination and laboratory data on the same people.

This is an invited submission to JPMedAI (Journal of Precision Medicine and Artificial Intelligence), a new, open-access, peer-reviewed journal currently assembling its inaugural issue. Article processing charges are waived during this launch phase, so there is no cost to you to publish. This suits researchers who value a genuine, citable publication and hands-on methodological support as much as, or more than, the cash fee, which reflects that context.

THE DATASET

Training dataset: NHANES continuous cycles, fully open, no registration.
External validation: temporal holdout using non-overlapping NHANES cycles, a genuinely independent national sample rather than a random split of the same years.

The angle: a great deal of prediction work has already been done on NHANES, so a model that restates a well-worn result will not stand out. We have a specific direction in mind, in the broad area of prospective risk stratification, that is less worked than it first looks. We have deliberately not fixed the exact question or endpoint here. In your application, propose your own take, tell us the outcome you would target and why, and be ready to ground it in a genuine gap you have checked. We refine the precise design together with the researcher we select.

GENERAL EXPECTATIONS FOR THE RESEARCH FELLOW

Use the survey's complex design correctly, meaning strata, primary sampling units and weights carried through the analysis, not treated as a simple random sample.
Report according to TRIPOD+AI (2024), the AI/ML-specific extension of TRIPOD, citing the original TRIPOD statement as the foundational reference. Use SHAP or an equivalent feature-attribution method for explainability.

Report discrimination, calibration as plots rather than a slope and intercept in the text, and decision-curve analysis against a clinically meaningful comparator, such as an existing risk score or a simple guideline rule, not against "no model" alone.
Test and report performance across demographic subgroups, including by income, insurance type and race or ethnicity, rather than a single pooled metric.

Before finalising the question, complete a focused, dated literature search on your exact endpoint and model, documenting the closest two or three prior papers and your specific point of difference.
Write the work up as a scientific manuscript, 3,000 to 6,000 words, up to 8 figures, structured as a 300-word structured abstract, keywords, introduction, methods, results, discussion, conclusion, and Vancouver-style references.

WHAT YOU'LL GET

An invited submission pathway to JPMedAI, with direct editorial engagement rather than a slot in a large backlog.
Hands-on methodological and editorial support to bring your first draft to a publishable, TRIPOD+AI-compliant standard.
Named byline, full citation credit, ORCID-compatible publication metadata, and a PDF for your portfolio, CV or dissertation.
A $100 honorarium on completion. The primary value on offer is the invited, mentored publication itself.
First-mover advantage on a specific, clinically useful question.

WHAT THIS IS NOT

This is not content writing, and it is not ghost-writing. You are an independent researcher producing original work under your own name.
The editorial process exists to hold scientific standards, not to change your conclusions or direct your inquiry.

IDEAL CANDIDATE PROFILE

Two kinds of applicant are welcome....
- Early-career researchers with at least one preprint or publication who want a fast, well-supported second one.
- Strong machine-learning or data-science practitioners without a clinical publication record who want to break into applied health research; for this group our support includes guidance on clinical framing, endpoint selection and TRIPOD+AI norms.

Useful but not required: background in epidemiology, biostatistics or survey-weighted analysis, and familiarity with calibration, decision-curve analysis or explainability methods.

Sharing the work with one or two collaborators who will be named co-authors is welcome.

IF YOU ARE INTERESTED

Answer the questions below and submit with your cover letter.

Link to a preprint, published paper, GitHub repo or portfolio demonstrating relevant work, in place of a CV; the work sample tells us what we need to know.
Write a short paragraph proposing your own research angle using the dataset. It is fine if it is rough; we will refine it together. If you can, name the closest prior paper you are aware of and one way your work would differ.
One or two sentences on your relevant background and how you would approach external validation for this data type.
Why does this specific opportunity interest you, beyond the honorarium?
Your availability and expected timeline. Typical engagement is 2 to 4 weeks from data access to a submitted draft, with one editorial round.

# 1. What we need from you

### Must have

**Backend (primary)**
- Strong TypeScript. Production Express (or equivalent Node backend) — middleware, auth, validation, error handling, background workers.
- **Real database design skill.** Not "I've used an ORM." You should be able to argue normalized-vs-JSONB with reasons, write migrations that are reversible, index for the query you actually run, and reason about transactions and idempotency. We run Postgres with Drizzle + hand-written SQL, one schema per module, ~190 tables. If you can't defend a schema decision, this role will hurt.
- Comfort with async job systems (BullMQ / ARQ / SQS-style), retries, idempotency keys, and callback contracts between services.
- Python is not required on day one, but you must be willing to work in the FastAPI sidecar. If you're allergic to Python, this isn't the role.

**Frontend**
- Next.js App Router and React in production — server/client component boundaries, data fetching, caching pitfalls.
- React Query (or equivalent) for server state; comfort with Tailwind + a Radix/shadcn-style component system.
- Able to build a dense, data-heavy UI — virtualized tables, inline editing, PDF viewers — not just marketing pages.

**Information extraction + LLM (the differentiator)**
- You have shipped a document-extraction system to production, and you know that **OCR/layout comes before the LLM**. Experience with Textract, or an equivalent OCR/layout stack, and an opinion about what to feed the model: raw text, layout-preserved markdown, images, or all three.
- **Structured output** as your default. Pydantic/Zod schemas as the contract, tool-use or JSON mode, and a considered answer to what happens when validation fails.
- **Prompt caching** — not the marketing version. You should be able to explain the prefix rule, why cache-write costs more than a normal input token, why a 5-minute TTL changes how you order your fan-out, and how to read `cache_read_input_tokens` to prove the cache is working.
- **Prompt engineering with the Anthropic API specifically** — system-vs-user placement, XML-tagged context, multimodal messages, `max_tokens` and streaming thresholds, and the difference between Bedrock and first-party Anthropic auth.
- **Evaluation instincts.** You've built a golden set, measured a change, and rolled back a prompt because the numbers said so.
- Token-cost awareness as a first-class engineering concern.

### Nice to have

- AWS: Bedrock, Textract, S3, SQS, EC2, RDS.
- Python/FastAPI, SQLAlchemy async, Alembic, ARQ.
- PDF internals — PyMuPDF, pdfjs, pdf-lib, rasterization, coordinate spaces.
- Docker Compose, pm2, nginx path-routing; you can deploy your own work to staging.
- CASL / policy-based authorization; multi-tenant RBAC.
- Construction domain knowledge (CSI MasterFormat, submittals, RFIs, SOV, pay applications) — genuinely valuable, but we can teach it.

### How we work — read this before applying

- **Verify before you claim.** "It should work" is not a status. We check staging, read the logs, and quote the output.
- **Docs are part of done.** A merged feature updates the relevant wiki standard and the sitemap.
- **Separate authoring from review.** You don't approve your own work.
- **Local-first.** Verify locally, then deploy deliberately. No surprise staging pushes.
- **No placeholder completions.** A `TODO`, a skipped test, or an unimplemented branch is a blocker to be reported, not evidence of progress.

---

## 2. Stack summary

| Layer | Tech |
|---|---|
| Frontend | Next.js (App Router), React, TypeScript, React Query, Tailwind, shadcn/Radix, Zustand, pdfjs/react-pdf, recharts, reactflow |
| Backend | Node/TypeScript, Express, Drizzle ORM, Zod, Jest + Supertest; Python/FastAPI sidecar with SQLAlchemy (async), Alembic, Pydantic, ARQ |
| Data | PostgreSQL (RDS), Redis, S3 |
| AI | Claude via AWS Bedrock (Sonnet + Haiku), `instructor`, Anthropic prompt caching, AWS Textract, embeddings/Pinecone, LangSmith tracing; Gemini + OpenAI as alternate providers behind one client interface |
| Infra | Docker Compose, pm2, nginx, EC2, SQS/NATS, GitHub |

We are looking for a reliable B2B Lead Generation Specialist to help identify decision-makers in US electrical distributors, builder supply companies, and electrical OEMs.

The goal is not bulk lead collection. We need accurate contact mapping of people involved in supplier selection, sourcing, category decisions, and vendor management.

Deliverables:

* Relevant contacts by company
* Job title and LinkedIn profile
* Verified business email
* Company/role start date (if available)

Ideal candidates have experience with industrial B2B research, LinkedIn sourcing, and decision-maker identification. Quality, accuracy, and attention to detail are more important than volume.

Overview

We're looking for an experienced AI Consultant to help shape our AI strategy and identify practical opportunities to introduce AI agents and intelligent automation across our business. This is a consulting role focused on architecture, planning, and technical guidance rather than full-time implementation.

You'll work closely with our team to evaluate existing workflows, recommend the right AI approach, and help us make informed decisions about agentic AI, RAG, LLMs, and automation.

Responsibilities

* Review our current business processes and identify opportunities for AI-driven automation.
* Recommend the most suitable architecture, including single-agent, multi-agent, or hybrid AI systems.
* Design a high-level roadmap for implementing Retrieval-Augmented Generation (RAG) using our internal knowledge and documentation.
* Advise on selecting the most appropriate LLMs, such as Claude, OpenAI, Gemini, or other models, based on our requirements.
* Evaluate any existing AI solutions and provide recommendations for improving performance, scalability, and reliability.
* Explain technical concepts and trade-offs in clear, business-friendly language to help stakeholders make informed decisions.
* Provide best practices around AI governance, security, cost optimisation, and long-term maintainability.

Required Experience

* Proven experience designing or consulting on AI agent or agentic AI solutions.
* Strong understanding of multi-agent architectures, orchestration frameworks, and AI workflows.
* Hands-on experience with Claude, OpenAI, Gemini, or similar enterprise LLMs.
* Practical experience designing and implementing RAG systems using vector databases and knowledge retrieval techniques.
* Ability to communicate complex technical concepts clearly to non-technical stakeholders.
* Experience advising businesses on AI strategy, architecture, and implementation planning.

Nice to Have

* Experience with MCP (Model Context Protocol), LangGraph, CrewAI, AutoGen, OpenAI Agents SDK, or similar frameworks.
* Knowledge of automation platforms such as n8n, Make, or Zapier.
* Familiarity with cloud platforms including AWS, Azure, or Google Cloud.
* Experience deploying production-grade AI solutions for internal operations or customer-facing products.

Deliverables

* Assessment of current workflows and AI opportunities.
* Recommended AI architecture and technology stack.
* AI implementation roadmap with prioritised use cases.
* RAG strategy and knowledge management recommendations.
* LLM selection guidance with cost, performance, and capability comparisons.
* Actionable recommendations and next steps for implementation.

To Apply

Please include:

* A brief overview of similar AI consulting or architecture projects you've worked on.
* Examples of multi-agent or RAG solutions you've designed or implemented.
* Your preferred AI frameworks and tools.
* A short explanation of how you would approach advising a business that is beginning its AI adoption journey.

# AI Data Scientist / AI Developer with Call Support (Full-Time)

## Job Overview

We are looking for a highly motivated **AI Data Scientist / AI Developer** to join our team full-time. This role combines AI development with client communication, so you must be comfortable participating in technical calls and collaborating with stakeholders in fluent English.

We are seeking candidates based in **India or Pakistan** who are available to start immediately.

## Responsibilities

* Develop and maintain AI/ML solutions and data-driven applications.
* Participate in client meetings and technical discussions in English.
* Understand business requirements and translate them into technical solutions.
* Collaborate with remote team members across different time zones.
* Provide technical support during client calls when needed.
* Deliver high-quality work while meeting project deadlines.

## Requirements

* Based in **India or Pakistan**.
* Strong experience in **AI, Machine Learning, Data Science, or LLM-based applications**.
* **English proficiency at C1 or C2 level** (both spoken and written).
* Excellent verbal communication and presentation skills.
* Available to work during **PST or EST business hours**.
* Reliable high-speed internet connection.
* Quiet, professional home office suitable for video calls.
* Available to **start immediately**.
* Available to work **full-time (40 hours/week)**.
* Self-motivated, reliable, and responsive.

## Preferred Qualifications

* Experience with Python and AI frameworks such as TensorFlow, PyTorch, LangChain, or OpenAI APIs.
* Experience building LLM applications, RAG systems, AI agents, or data pipelines.
* Experience working with international clients.

## Compensation

* **$10 USD/hour**
* Full-time, long-term opportunity for the right candidate.

## To Apply

Please include the following in your proposal:

1. A brief introduction.
2. Your AI/Data Science experience.
3. Examples of AI projects or GitHub portfolio (if available).
4. Your English proficiency level.
5. Your current location.
6. Your earliest available start date.

We seek a data scientist or someone who has skills in to review a specific protocol and:

- structure database
- add logic
- export as ODBC compatible spreadsheet
- provide feedback to programmers on how to implement on server
- provide extraction chunks of relevant data for RAG style searches

IMPORTANT NOTE:
A NDA must be signed with matching photo and signature on ID to engage in this project.

Successful applicant will be offered other protocols to conduct same task as separate milestone.

We encourage recent students graduating and data scientists to apply.

We DO NOT seek programmers to do this as it is a specific data science project, please respect our wishes - look forward to getting your feedback

I need hands-on guidance getting a full CUDA environment installed on my Windows 11 workstation so I can run and train convolutional neural networks reliably on an RTX 4090. The NVIDIA driver is already current, but CUDA, cuDNN and the deep-learning framework still have to be matched correctly, then tied into a clean Python environment (probably via Anaconda or venv).

Your job is to walk me through the entire process in a live session, troubleshoot any conflicts, and leave me with a ready-to-use project folder that can import the standard CNN examples without errors. I want to see nvidia-smi registering the GPU, the framework recognising it and a short sample model completing a forward pass.

Deliverables:
• Interactive support until the environment is fully operational.
• A concise summary of the exact versions and commands used, so I can replicate the setup later.

If you’ve already tamed CUDA on a 40-series card under Windows 11, this should be straightforward—we just have to make it happen on my machine.

Skills: C Programming, Python, Script Install, Troubleshooting, CUDA, Machine Learning (ML), Data Science, Deep Learning, Remote Quality Audit, AI Development

The Company:

Dimensional Strategies Inc. (DSI), is a premier Microsoft partner located in the GTA, Ontario, Canada. Our customer base extends across Canada and the contiguous US states. DSI’s primary goal is to help our customers execute their business processes more efficiently and effectively via investments in applications and analytics. Core practices include Cloud Infrastructure, Application Development, Data Engineering & Analytics, and Managed Services.

The Position:

Participate in the end-to-end execution and delivery of our data platform projects that focus on ingesting and curating data for analytical consumption. You will be using DSI's flagship in-house metadata-driven framework for Microsoft Fabric that is based on Medallion Lakehouse architecture.

You will work with the DSI data engineering team and use data engineering specifications to create production-ready data pipelines across Bronze (ingestion), Silver (cleansing), and Gold (curation) layers.

This is a hands-on delivery role - not architecture, not requirements gathering, not custom data engineering or data science.

What You'll Do

· Bootstrap and provision new customer environments in Microsoft Fabric
· Author and maintain YAML-based pipeline configurations for all data layers
· Deploy configurations and artifacts to Microsoft Fabric via CI/CD pipelines
· Configure data ingestion from various data sources including REST API, file sources, and SQL databases
· Define merge strategies, column transformations, and data quality rules
· Validate pipeline outputs by running notebooks and verifying table data in Fabric
· Debug pipeline failures using logs, tracing config errors to resolution
· Manage multiple environments (dev, test, production) across customer projects
· Collaborate with the DSI data engineers that define the specs for the solutions

What You Won't Do

· Gather requirements or collaborate with customers
· Write or modify the data engineering specification
· Perform custom data engineering or research in Microsoft Fabric
· Perform business analysis
· Design solution architecture from scratch

Primary Qualifications

· Multiple years of dedicated data engineering experience
· Strong SQL skills with Spark SQL and T-SQL - window functions, CTEs, aggregations, CASE expressions
· Experience with Lakehouse architecture in Microsoft Fabric including Delta tables, and semi-structured file formats
· Solid understanding of Medallion architecture (Bronze/Silver/Gold) or equivalent layered data patterns
· Experience with REST API integrations - authentication, pagination, JSON response handling
· Familiarity with merge/upsert patterns (SCD Type-1, Type-2, overwrite, append)
· Proficiency with Git - branching, pull requests, merge conflict resolution
· Comfortable reading and authoring YAML configurations
· Understanding of Apache Spark concepts — partitions, lazy evaluation, actions vs transforms

Nice to have Qualifications

· Exposure to CI/CD pipelines (Azure DevOps or GitHub Actions)
· Basic Python reading ability (not writing production code, but can trace logic)
· Familiarity with expression languages like Jinja2
· Experience with Python scripts execution
· Experience with PowerShell scripts execution
· KQL / Kusto query experience for log analysis
· Databricks or Azure Synapse experience

Work Environment

· Fully remote with flexible time management
· Initially supervised by DSI team members, progressing to semi-autonomous and then fully autonomous delivery responsibilities
· Multi-customer project rotation
· Work in a mature well-documented Microsoft/GitHub enterprise environment using established patterns and templates

Project Overview

I’m looking for an experienced developer to build an AI-powered trading assistant that helps traders analyze the market and make more informed trading decisions.

The goal is to create a tool that combines technical analysis, market trends, price action, and relevant news into one intelligent platform. This is not an automated trading bot or a system that guarantees profits. It is designed to provide traders with insights, trade ideas, and educational explanations to support their decision-making.

Key Features

* AI-assisted market analysis
* Buy/Sell trade suggestions based on multiple market factors
* Trend analysis across multiple timeframes
* Support & Resistance detection
* Candlestick and price action recognition
* Entry, Stop Loss, and Take Profit suggestions
* Confidence score for each trade idea
* News and market sentiment analysis
* Clear explanations for why each signal is generated
* Clean, intuitive, and user-friendly interface

Requirements

* Experience with AI and machine learning
* Strong software development skills (frontend & backend)
* Experience working with financial data and APIs
* Knowledge of technical analysis and trading concepts is a plus

Long-Term Vision

This project is intended to grow into a subscription-based platform that can help traders of all experience levels make better-informed decisions. I’m looking for a developer interested in building a scalable, reliable, and high-quality product with the potential for long-term collaboration as new features are added.

Budget Pay is negotiable

We are seeking a skilled freelancer to assist in building an AI data hub center in the Northeast. The project involves developing a comprehensive platform for data management and AI integration. The ideal candidate will have experience in AI technologies and data infrastructure development. This is a part-time role with a short-term engagement.

I have a rich set of customer data—demographic profiles, detailed transaction histories, and on-site behavioral logs—that I need transformed into clear, actionable insights. My goal is to understand who my customers are, how they buy, and what drives their engagement so I can sharpen marketing, retention, and product decisions.

Here’s the flow I have in mind. You will first explore, clean, and merge the three data sources, making sure key identifiers align. From there I expect descriptive statistics, visual exploration, and segment discovery. Once the groundwork is solid, I’m interested in deeper analysis such as cohort trends, lifetime value estimates, or predictive models if the patterns warrant it. I work comfortably with Python (pandas, scikit-learn), SQL, or R, and I’m happy to review results in Jupyter, a clean slide deck, or an interactive Tableau/Power BI dashboard—whatever best showcases the story in the numbers.

Deliverables:
• A well-commented notebook or script that reproduces all cleaning and analysis steps
• A concise report or dashboard highlighting key findings, charts, and recommendations
• The final, processed dataset ready for future use

Acceptance criteria:
• All three data types—demographic, transaction, and behavioral—are integrated without loss of essential fields
• Insights are supported by clear visuals and, where relevant, statistical significance or model metrics are provided
• Code runs end-to-end on my sample environment with no missing dependencies

If you’re experienced in data wrangling and storytelling, I’m ready to share a sample extract so you can gauge complexity. Let’s turn raw customer records into decisions I can act on.

Skills: Python, SQL, Statistics, Statistical Analysis, SPSS Statistics, Data Science, Tableau, Data Visualization, Data Analysis, Power BI

I need a custom algorithmic trading application that executes intraday positions exclusively in the stock market. The core logic will blend three data streams in real time—classic technical indicators (moving averages, RSI, VWAP, etc.), key fundamental metrics released during the session, and live news-sentiment scores—so that the system can open, scale, and close trades within the same day.

The workflow I have in mind is:

• Data ingestion: pull tick-level price feeds and corporate fundamentals, then pipe breaking-news headlines through a sentiment engine.
• Signal engine: combine weighted signals from those three sources into clear long/short triggers.
• Risk and execution: apply position-sizing rules, hard/soft stops, and route orders automatically to my preferred brokerage API.
• Monitoring & override: a lightweight dashboard or CLI where I can watch positions, pause the bot, or adjust parameters on the fly.
• Post-trade analytics: export logs for daily review, including P&L, slippage, and hit rate.

Please build the code in a widely supported language (Python preferred, but I’m open to C++ or Java if it boosts performance) and structure it so I can tweak indicator weights without rewriting functions. A walk-forward back-test covering at least one year of 1-minute data is required before we go live.

Deliverables:
1. Fully commented source code with installation instructions.
2. Config file or UI panel for parameter changes.
3. Back-test report (PDF or Jupyter Notebook) showing key metrics.
4. Brief hand-off session to walk me through deployment.

If something here needs clarification, just let me know; otherwise, feel free to suggest libraries or broker APIs that fit this day-trading setup.

Skills: Java, Python, Financial Analysis, Data Science, Data Analysis, API Development, Backtesting, Sentiment Analysis

We're a B2B marketing agency looking for a contract data scientist to support ongoing analytics work for manufacturer clients with dealer-based distribution (boats, RVs, commercial vehicles).

This first project is a paid test. We want to see how you think before we scope a longer relationship.

The setup:
We'll share a real client report built for a marine (boat) client, including foot traffic data, geo-based dealer comparisons, and web analytics lift analysis. You'll get:
- The original brief/goals for the report
- The finished v1 output another data scientist produced
- The underlying data used to build it

What we need from you:
A written assessment covering:
- How you'd approach reproducing this analysis from scratch
- What's solid vs. what you'd change or question in the existing methodology
- What it would take to extend it (additional data sources, more granular breakouts, better lift/attribution methods)
- Rough time estimate to repeat vs. extend

We're not asking you to build anything yet. This is about how you reason through messy, partial marketing data (think: limited pixel tracking, inconsistent UTMs, no direct sales data) and where you'd push back or ask better questions.

Good fit if you:
- Have experience with marketing/media analytics (GA4, paid media platforms, geo/foot traffic data like Placer.ai)
- Are comfortable with imperfect data and can reason around gaps instead of stalling on them
- Can communicate technical tradeoffs clearly to non-technical stakeholders

Structure: Fixed price for this assessment. If it's a strong match, we're looking to bring someone on for ongoing ad hoc contract work.

The Company:

Dimensional Strategies Inc. (DSI), is a premier Microsoft partner located in the GTA, Ontario, Canada. Our customer base extends across Canada and the contiguous US states. DSI’s primary goal is to help our customers execute their business processes more efficiently and effectively via investments in applications and analytics. Core practices include – Cloud Infrastructure, Application Development, Data Engineering & Analytics, and Managed Services.

The Position:

Participate in the end-to-end execution and delivery of our data platform projects that focus on ingesting and curating data for analytical consumption. You will be using DSI's flagship in-house metadata-driven framework for Microsoft Fabric that is based on Medallion Lakehouse architecture.

You will work with the DSI data engineering team and use data engineering specifications to create production-ready data pipelines across Bronze (ingestion), Silver (cleansing), and Gold (curation) layers.

This is a hands-on delivery role - not architecture, not requirements gathering, not custom data engineering or data science.

What You'll Do

· Bootstrap and provision new customer environments in Microsoft Fabric
· Author and maintain YAML-based pipeline configurations for all data layers
· Deploy configurations and artifacts to Microsoft Fabric via CI/CD pipelines
· Configure data ingestion from various data sources including REST API, file sources, and SQL databases
· Define merge strategies, column transformations, and data quality rules
· Validate pipeline outputs by running notebooks and verifying table data in Fabric
· Debug pipeline failures using logs, tracing config errors to resolution
· Manage multiple environments (dev, test, production) across customer projects
· Collaborate with the DSI data engineers that define the specs for the solutions

What You Won't Do

· Gather requirements or collaborate with customers
· Write or modify the data engineering specification
· Perform custom data engineering or research in Microsoft Fabric
· Perform business analysis
· Design solution architecture from scratch

Primary Qualifications

· Multiple years of dedicated data engineering experience
· Strong SQL skills with Spark SQL and T-SQL - window functions, CTEs, aggregations, CASE expressions
· Experience with Lakehouse architecture in Microsoft Fabric including Delta tables, and semi-structured file formats
· Solid understanding of Medallion architecture (Bronze/Silver/Gold) or equivalent layered data patterns
· Experience with REST API integrations - authentication, pagination, JSON response handling
· Familiarity with merge/upsert patterns (SCD Type-1, Type-2, overwrite, append)
· Proficiency with Git - branching, pull requests, merge conflict resolution
· Comfortable reading and authoring YAML configurations
· Understanding of Apache Spark concepts — partitions, lazy evaluation, actions vs transforms

Nice to have Qualifications

· Exposure to CI/CD pipelines (Azure DevOps or GitHub Actions)
· Basic Python reading ability (not writing production code, but can trace logic)
· Familiarity with expression languages like Jinja2
· Experience with Python scripts execution
· Experience with PowerShell scripts execution
· KQL / Kusto query experience for log analysis
· Databricks or Azure Synapse experience

Work Environment

· Fully remote with flexible time management
· Initially supervised by DSI team members, progressing to semi-autonomous and then fully autonomous delivery responsibilities
· Multi-customer project rotation
· Work in a mature well-documented Microsoft/GitHub enterprise environment using established patterns and templates

Looking for an experienced statistician or data scientist to work with us from mid-October to mid-November 2026. The role involves designing data models and collaborating on related projects. The ideal candidate should have a strong background in data modeling, data analysis, statistics and preparing system design specifications. Availability during the specified time frame is essential.

We are looking for an experienced full-stack AI developer proficient in RAG and LLM integration for a short-term task. You will be responsible for auditing, debugging, and improving the retrieval accuracy of our current implementation. We are specifically seeking someone who has shipped real production RAG systems and successfully resolved latency and context precision issues.

You must have strong expertise with VectorDBs (Pinecone, ChromaDB), embedding-based retrieval, hybrid search, re-ranking, and advanced prompt engineering. Our current setup uses Python, OpenAI's API, and a vector database - the issues are mainly around context relevance, retrieval precision, and inconsistent or hallucinated answers when processing documents with highly similar content.

To apply, please share a concrete example of a production RAG system you have developed, along with your immediate availability to start this week.

I developed an AI-powered Skin Specialist application that uses computer vision and Generative AI to analyze skin images and provide AI-generated insights. The application also integrates voice interaction and text-to-speech capabilities, creating a more interactive user experience. Technologies used include Python, Groq Vision, Deepgram, Gradio, PIL, and API integration.

Skills: Python, Machine Learning (ML), Data Science, API, Computer Vision, Deep Learning, Gradio, Generative AI, Large Language Models (LLMs)

About Us
We are a Canadian-built technology company operating in the CRM and marketing technology space. Our platform helps businesses manage, analyze, and activate customer data at scale. We build solutions that process millions of records, enabling organizations to improve customer engagement, marketing automation, and data-driven decision-making.

As we expand our AI capabilities, we are looking for talented Software Engineers to help shape the future of AI-powered development tools and enterprise software solutions.


Role Overview

We are seeking experienced Software Engineers to support cutting-edge AI model training and evaluation projects. You will work closely with AI researchers and engineering teams to improve large language models (LLMs) by creating high-quality coding datasets, evaluating AI-generated solutions, and developing benchmarks for software engineering capabilities.

This role is ideal for engineers who have designed, built, and shipped high-impact, production-grade systems in fast-moving technology environments.

Experience at companies such as Stripe, Airbnb, Cloudflare, Datadog, Coinbase, or similar high-growth engineering organizations is highly valued.

Real-world engineering experience, technical depth, and demonstrated impact matter most.

Responsibilities
Create and curate high-quality software engineering datasets for LLM training and benchmarking.
Develop, review, and improve code examples across:
Python
C/C++
Rust
Go
Java
JavaScript / ReactJS
Evaluate AI-generated code for:
Correctness
Performance
Scalability
Reliability
Maintainability
Analyze systems-level, backend, and infrastructure code to identify:
Bugs
Performance bottlenecks
Design issues
Improvement opportunities
Develop evaluation frameworks and verification methods to measure AI coding capabilities.
Design automated approaches to validate software engineering solutions.
Collaborate with AI researchers and engineers to improve AI-driven coding systems.
Evaluate software engineering workflows across the full development lifecycle:
Prototyping
Architecture design
API design
Production implementation
Testing and experimentation
Monitoring and operational maintenance
Typical Day

Your work may include:

Reviewing AI-generated code and providing detailed technical evaluations.
Creating challenging software engineering problems and reference implementations.
Improving benchmark datasets for coding agents and AI assistants.
Designing verification workflows to measure code quality and reliability.
Partnering with research teams to identify model limitations and improve AI performance.
Required Qualifications
3+ years of professional software engineering experience.
Strong background in backend engineering, infrastructure, distributed systems, or systems programming.
Strong proficiency in one or more of the following:
Python
C/C++
Rust
Go
Java
JavaScript / ReactJS
Experience designing and deploying scalable, production-grade software.
Strong understanding of:
Software architecture
System design
Debugging
Code review practices
Engineering best practices
Ability to communicate complex technical reasoning clearly through written evaluations and documentation.
Preferred Qualifications
Experience building large-scale distributed systems or developer infrastructure.
Experience with performance optimization and low-level programming.
Experience working with:
AI/ML systems
Large Language Models (LLMs)
AI coding assistants
Developer productivity tools
Experience reviewing complex enterprise production codebases.
Background working at high-growth technology companies or engineering-intensive environments.
Experience with CRM, marketing technology, customer data platforms, or enterprise SaaS is a plus.

AI Engineer to Build a Personalized Finance Podcast Intelligence Assistant

We're looking for an experienced AI engineer to build a private system that turns long-form finance podcasts into concise, actionable investment insights.

The goal is not simply summarization, we want an AI research assistant that understands recurring themes across dozens of podcasts and produces high-quality investment intelligence.

Responsibilities
Pull podcast episodes automatically (YouTube, Spotify, RSS)
Transcribe audio using Whisper or similar
Break transcripts into logical sections
Use GPT-5 / Claude to summarize key ideas
Extract:
Investment theses
Company mentions
Macro themes
Bull vs. bear arguments
Actionable takeaways
Risks
Compare ideas across multiple podcasts
Detect recurring themes over time
Store everything in a searchable knowledge base (Notion or similar)
Deliver a daily digest by email or Slack
Nice-to-have

Experience with:

OpenAI API
Claude API
LangChain or LangGraph
n8n
Make.com
Python
Whisper
Pinecone
Weaviate
Supabase
Notion API
Retrieval-Augmented Generation (RAG)
MCP servers
Agentic AI workflows
End Goal

The finished product should function like a personal investment research analyst.

Example daily output:

Top 5 ideas mentioned today
Highest-conviction stock picks
Consensus themes across podcasts
New macro risks emerging
Contrarian viewpoints
Companies mentioned most frequently
Actionable investment opportunities

The system should become smarter over time by remembering previous episodes and connecting related ideas rather than treating every podcast independently.

Prepare Code and Plots with R studio, interpret coefficients, p value and conduct bivariate analysis, simple linear regression and multiple linear regression controlling for confounder.

Research Question: To what extent does the location of a restaurant influence the confidence levels of written allergen information received by individuals with food hypersensitivity?

Null Hypothesis (H0): There is no difference in confidence levels regarding written allergen information based on restaurant location.

Alternative Hypothesis (H1): There is a difference in confidence levels regarding written allergen information based on restaurant location.

Alternative Hypothesis (H2): There is a differential effect of restaurant location on confidence levels based on whether the individual has a food allergy.

To do: conduct univariate and bivariate descriptive analyses of variables obtained from food and you 2:wave 4 uk data service;
run at least two linear regression models (one simple linear regression model with just your dependent and independent variables and at least one multiple linear regression model controlling for a confounder);
interpret your results substantively.

Dependent Variable: Confidence in Written Information, column 66 - very confident, fairly confident, not confident, not confident at all - "allerstaff2_a"
Independent Variable: Location - column 6 - urban/rural - "urban"
Confounding Variable: Allergy Demographic - column 26, have allergy? 1=yes, 2=no, 3&4 = NA. "foodreac"

Skills: Statistics, R Programming Language, Statistical Analysis, SPSS Statistics, Data Science, Data Visualization, Data Analysis, Statistical Modeling, Linear Regression, Regression Analysis

I am an MD Psychiatry postgraduate with a finished protocol, defined objectives and a fully de-identified master Excel sheet. I now need an experienced biostatistician to turn that raw file into publication-ready results.

You may work in SPSS or R—either is fine—so long as every step is fully documented. The dataset contains some missing values; I expect you to diagnose, report and address them before any modelling begins.

Key deliverables
• Dataset check-up and thorough cleaning
• Objective-wise statistical analysis plan
• Descriptive summaries plus appropriate inferential tests, including regression where justified
• Verification of assumptions and treatment of any missing data or outliers discovered later
• Publication-ready tables, figures and a clear Statistical Methods section
• Complete SPSS/R output and annotated syntax
• Draft Results section with concise interpretation
• Up to two revisions after my guide’s feedback

Before we start, please review the protocol and variable list and give me a brief outline of the tests and tables you propose.

When you reply, attach a detailed project proposal that highlights your qualifications and prior medical-research work, and quote a single fixed fee; I have allocated roughly ₹3,000–₹5,000 for the package.

Skills: Statistics, Statistical Analysis, SPSS Statistics, Data Science, Data Analysis, Statistical Modeling, Data Management, Biostatistics

Role Overview:
The position focuses on building and validating golden datasets for an AI/LLM-based Natural Language Query (NLQ) Evaluation Framework. You'll be responsible for designing synthetic data generation pipelines, developing SQL-based QA templates, validating datasets, and creating schema documentation while collaborating with domain experts and engineering teams.
Key Responsibilities:
Design and implement synthetic data generation pipelines using Python, NumPy, Faker, and pandas.
Generate realistic datasets with statistical distributions and controlled data imperfections.
Validate datasets using DuckDB and maintain dataset versioning.
Develop Jinja2 SQL templates and verify expected answers against reference SQL.
Collaborate with Domain SMEs to create QA datasets and support LLM evaluation.
Create ER diagrams, DDL scripts, Data Dictionaries, and CSV Header Specifications.
Required Skills:
Strong Python programming with NumPy, pandas, and SQL
Experience with synthetic data generation and data validation
Knowledge of DuckDB, Jinja2, and Git
Understanding of database design, schema documentation, and SQL development
Exposure to AI/LLM evaluation frameworks is an added advantage

We are seeking an experienced consultant to help us analyze and implement AI solutions aimed at improving our operational efficiency. The ideal candidate will have a strong background in AI technologies and a proven track record of enhancing business processes through innovative solutions. You will work closely with our team to identify areas for improvement and develop actionable strategies that leverage AI for maximum impact. If you are passionate about driving efficiency through technology, we want to hear from you!

update org chart you previously worked on last year.

About the Role
We're building out our business intelligence capability on a new Metabase instance integrated with Odoo, and we're looking for someone to design and build dashboards that turn our Odoo data into clear, actionable insights for the team.

What You'll Do

Design, build, and maintain dashboards in Metabase
Write and optimize SQL queries against our Odoo database
Translate business questions from different departments into meaningful metrics and visualizations
Ensure data accuracy, consistency, and performance across reports
Collaborate with stakeholders to understand reporting needs and iterate quickly

What We're Looking For

4+ years of experience as a Data Analyst or Data Scientist
Strong SQL skills and hands-on experience building dashboards (Metabase, Power BI, Tableau, Looker, or similar)
Solid understanding of data modeling and how to structure clean, reliable reports
Ability to work independently and communicate insights clearly to non-technical stakeholders

Nice to Have

Familiarity with Odoo and its data structure; since our dashboards run on Odoo data, prior experience means you won't need us to walk you through the schema, models, or table relationships
Experience with ETL processes or data pipelines
Python or other scripting for data transformation

We are seeking an AI Engineer to review and enhance our codebase. The ideal candidate will have experience in AI and machine learning, with a strong understanding of code optimization and best practices. Responsibilities include reviewing existing code, identifying areas for improvement, and implementing changes to enhance performance and efficiency. Familiarity with microcontrollers and embedded systems is a plus.

I'm looking for an experienced AI consultant to discuss an AI project and provide expert guidance on the best approaches and technologies. The consultant should have a strong understanding of AI systems and be able to offer strategic advice on project development. The role involves analyzing project requirements and suggesting innovative solutions to enhance AI integration.

I'm looking for an experienced AI engineer to review my existing AI application during a one-hour consultation. The focus will be on code review and architecture, providing insights and recommendations for improvement. The ideal candidate should have a strong background in AI and software development, with the ability to analyze complex systems and provide actionable feedback.

We're building an AI system that reads content and automatically determines:


- Whether it contains a paid promotion (yes / no)
- What type of endorsement it is, based on our classification spec
- Which brand it refers to: matched against our existing brand database, with a suggested new brand name when there's no match

We already have a labeled set of examples across the different endorsement types that you can use for training and evaluation.

# Quantitative ML Engineer — Tick Data, Feature Engineering and Predictive Modeling

We have an existing systematic trading strategy and execution system. We are seeking a hands-on quantitative ML engineer to build and manage the dataset and modeling pipeline that supports the strategy.

The selected candidate will take historical tick data and candidate trade events, construct an event-level ML dataset, engineer predictive features, and train models that score which opportunities are most likely to produce favorable continuation or returns.

Some existing data-processing components are written in Rust. Most ML research may be conducted in Python, but you must be comfortable working with or integrating into a Rust-based data pipeline. Direct Rust development experience is strongly preferred.

## Responsibilities

- Process and manage historical tick, trade, and quote data
- Build reproducible event-level modeling datasets
- Maintain and improve existing Python and Rust data pipelines
- Engineer market, microstructure, liquidity, regime, and security-relative features
- Research and integrate alternative data where it may add predictive value
- Define labels and prediction targets around the existing strategy
- Train and compare statistical and machine-learning models
- Perform walk-forward and out-of-sample validation
- Prevent look-ahead bias, target leakage, survivorship bias, and overfitting
- Monitor datasets, features, and model performance over time
- Deliver a scoring model that can be integrated into the existing system
- Clearly document the dataset, experiments, results, and limitations

## Required Experience

- Strong Python and machine-learning experience
- Financial time-series, tick-data, or high-frequency data experience
- Feature engineering for sequential or event-based data
- Time-aware model validation
- Building and maintaining reproducible data pipelines
- Ability to work independently and take ownership of the data and ML process

Experience with Rust, market microstructure, systematic trading, alternative data, Parquet, Polars, DuckDB, SQL, or cloud-based datasets is highly desirable.

## Project Structure

The trading strategy and execution platform already exist. You will not be responsible for inventing the strategy or building the brokerage execution system.

The work will be siloed. You will receive the relevant data, candidate-event records, prediction objectives, and enough information to understand the strategy’s decision context. Access to brokerage accounts, portfolio logic, and the complete production system is not required.

We will begin with a paid pilot using a limited historical period and security universe. A successful pilot may lead to an ongoing engagement managing the datasets, engineering additional features, researching alternative data, retraining models, and monitoring performance.

Individual freelancers are preferred over agencies.

## Application Questions

Please begin your proposal with “TICK DATA” and answer:

1. What experience do you have building ML datasets from tick or other high-frequency financial data?
2. How did you prevent leakage and validate the model over time?
3. What experience do you have with Rust-based data pipelines?
4. Have you evaluated or integrated alternative data into financial models?
5. How many hours per week can you consistently dedicate?